[R] Q re Linear Models

Joerg Evermann joerg.evermann at vuw.ac.nz
Wed Jul 16 22:56:47 CEST 2003


I apologize in advance if what I'm about to ask is trivial or has been 
answered before. In the latter case I would appreciate a pointer to the 
right list/location

I'm trying to model the following experimental design with groupedData 
and lme in R:

Subjects were measured on two tasks (continuous outcome variable is 
"Prob", tasks are coded as within-subjects factor "Domain"). Subjects 
are recruited from three different groups ("Group") and assigned to 
three different conditions ("Rules"). These all appear to be fixed 
between-subjects factors. The effect I'm after is that of "Rules".

Additionally, I've measured for each observation (for each subject for 
each task) four continuous variables ("SelfAssessment", "Usefulness", 
"Information" and "Ease") that I'd like to include as covariates. These 
of course may partially depend on the subject. Additionally, in a 
post-test I measure yet another continuous variable for each subject 

The problem I have is that I'm not sure how to properly include these 
continuous covariates in the groupedData formulas and the lme formulas.

Any help will be appreciated, either on the list of by email.

Many many thanks in advance.


More information about the R-help mailing list